Post on 22-Apr-2022
WHITE PAPER
Power Plant Model Validation (PPMV) with MATLAB and Simulink Graham Dudgeon PhD | MathWorks Principal Industry Manager Utilities amp Energy
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 2
Introduction
If you want to create a large-scale simulation of an electric grid that closely matches reality you gen-erally have to calibrate the individual components of that grid This process power plant model vali-dation (PPMV) can be performed with both offline step tests and online performance monitoring of grid events The main objective of PPMV is to
bull Find potential errors andor fixes in the model
bull Understand the sensitivity of parameters to potential model improvements
This task can be challenging particularly when required by technical regulations such as NERC Standard MOD-026
This paper covers a workflow for PPMV using MATLABreg and Simulinkreg with emphasis on online performance monitoring of grid events using phasor measurement unit (PMU) data It explores work-flows that include both manual adjustments and automated techniques A gas plant case study dem-onstrates how to
1 Replay measured data through your simulations
2 Gain insight into response discrepancies through both VF and PQ replay
3 Use engineering judgement and automated parameter sensitivity to assess and rank the influence of system parameters on system response
4 Fine-tune your system response using both manual adjustments and automated parameter estimation
The grid-connected gas plant model structure and grid event data were provided as part of the NASPINERC Technical Workshop on Model Verification Tools The model consists of the following components
bull Round rotor generator
bull General rotating exciter
bull GGOV1
bull PSS2A
bull Step-up transformer and series inductor
Replay Measured Data Through Your Simulations
The core attribute of PPMV is to replay measured data through a simulation model For offline step-tests data replay involves exciting the plant model through control signal measurements For online performance monitoring data replay involves exciting the plant model through physical measure-ments of voltage or current at the grid point-of-connection It is common to consider voltage (V) fre-quency (F) active-power (P) and reactive-power (Q) as the four measurements necessary to perform PPMV
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 3
Two data replay paradigms are available with these four measurements
1 PQ replay and compare simulated VF with measured VF
2 VF replay and compare simulated PQ with measured PQ
Figure 1 shows an example of the two data replay paradigms In this case the simulation model is cor-rupt and requires parameter modification A natural question that arises is which paradigm should be chosen to perform PPMV To obtain the most insight into system response and establish the most robust parameter estimation workflow the answer is both
Figure 1 VF replay with PQ comparison (top) and PQ replay with VF comparison (bottom)
Gain Insight into Response Discrepancies Through Both VF and PQ Replay
The first stage in PPMV is to gain insights into response discrepancies through manual adjustment of parameters You may have a list of ldquogo-tordquo parameters to adjust based on previous experience and can also gain additional insights through observing the attributes of a response discrepancy which can point to certain parameters requiring adjustment Figure 2 shows the results of PQ replay for the gas plant example As the simulated VF response does not match the measured VF response the model is corrupt and requires parameter adjustment The increased magnitude of frequency oscillation on the simulation indicates that generator inertia is too low Figure 3 shows the results of PQ replay with generator inertia increased The simulated oscillation now better fits the measured response
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 4
Figure 2 PQ replay for corrupt gas plant model
Figure 3 PQ replay with generator inertia increased
Figure 4 shows the results of VF replay for the gas plant example The reactive power response shows an underlying 180deg phase shift discrepancy in the simulated response This is indicative of the time-constants in two series washout filters in the power system stabilizer (PSS) requiring adjustment Figure 5 shows the results of VF replay with the PSS washout filter time-constants increased (Tw1 Tw2) The simulated reactive-power now better fits the measured response
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 5
Figure 4 VF replay for corrupt gas plant model
Figure 5 VF replay with PSS time-constants increased
Use Engineering Judgement and Automated Parameter Sensitivity to Assess and Rank the Influence of System Parameters on System Response
Further manual adjustments were made resulting in the response of Figure 6 for PQ replay and Figure 7 for VF replay Table 1 shows a comparison of parameter adjustments The results are certain-ly heading in the right direction and could be further improved manually but at this stage we will apply automated parameter sensitivity to assess and rank the influence of additional system parame-ters on system response
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 6
Figure 6 PQ replay for manually adjusted parameters
Figure 7 VF replay for manually adjusted parameters
Table 1 Manual parameter adjustments
Parameter Corrupt Value Manually Adjusted Value
pssTw1 10 60
pssTw2 10 60
pssKs1 150 250
govKturb 15 30
govKpgov 60 20
avrTf 50 10
genH 31 60
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 7
We use automated parameter sensitivity to assess parameters not evaluated during manual adjust-ments PQ replay and VF replay are used simultaneously for automated parameter sensitivity (PQVF replay) Figure 8 shows the model for PQ replay VF replay simply involves changing the data replay block indicated in red For PQVF replay the duplicate models share the same parameter values
Figure 8 Model for PQ replay (VF replay)
The objective function (f(x)) that is used to calculate sensitivity compares measured PQVF with sim-ulated PQVF The responses are normalized to mitigate scaling bias The objective function uses sum squared error (SSE)
Figure 9 shows a plot of the correlation of the plant parameters to the objective function While mag-nitude of correlation is certainly a factor for selecting a parameter sensitive parameters may be dis-counted based on engineering judgement For this case study we will select genTpd0 genTppd0 genXq and pssT4 for further consideration
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 8
Figure 9 Correlation of system parameter influence to objective function
Fine-Tune Your System Response Using Both Manual Adjustments and Automated Parameter Estimation
Following manual adjustment and automated parameter sensitivity you can apply automated param-eter estimation to fine-tune the response In this stage we have observed that solution convergence for PQVF replay is more robust and accurate than separate PQ replay or VF replay Parameter value ranges can be constrained in the automated parameter estimation task Both absolute and relative
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 9
constraints can be implemented Absolute constraints are typically adequate for control system parameters but relative constraints are required for generator parameters
Figure 10 shows a comparison of automated fine-tuning for both the manually selected parameters and the manually selected parameters in addition to the automatically selected parameters Bringing additional parameters to the parameter estimation task has improved the overall fit
Figure 10 Result of automated fine-tuning
It should be noted that the PPMV task should not end after automated fine-tuning You should assess the result and determine whether further manual adjustments can be made For example you can set parameters that do not change significantly back to their original values and compare responses If the result is comparable it may be more appropriate to stick with the minimum number of parameter changes Table 2 shows the final result of this case study where genTpd0 and genTppd0 did not change much during automated fine-tuning and were reset to their original values In this case the end result was slightly improved although this should not be regarded as a general outcome
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 1 0
Table 2 Case study results
Summary
In this paper we explored PPMV as applied to online performance monitoring of grid events using PMU data using a workflow that included both manual adjustments and automated techniques A gas plant case study demonstrated the following workflow steps
1 Replay measured data through your simulations
2 Gain insight into response discrepancies through both VF and PQ replay
3 Use engineering judgement and automated parameter sensitivity to assess and rank the influence of system parameters on system response
4 Fine-tune your system response using both manual adjustments and automated parameter estimation
With MATLAB and Simulink you can efficiently perform power plant model validation with auto-mated techniques The workflow provides insight and flexibility when addressing technical regula-tions such as NERC Standard MOD-26
Parameter Corrupt First-Pass Tuned
Second-Pass Tuned
Final
f(x) 9776 00872 00477 00472
pssKs1 150 2561 2564 2564
pssTw1 10 570 551 551
pssTw2 10 570 551 551
govKpgov 60 255 254 254
govKturb 15 30 30 30
avrTf 50 055 068 068
genH 31 60 60 60
genXq 13 13 177 177
genTppd0 025 025 023 025
genTpd0 50 50 505 50
pssT4 116 116 131 131
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 1 1
copy 2017 The MathWorks Inc MATLAB and Simulink are registered trademarks of The MathWorks Inc See mathworkscomtrademarks for a list of additional trademarks Other product or brand names may be trademarks or registered trademarks of their respective holders
93126v00 0517
Learn More About PPMV with MATLAB and Simulink
Power Plant Model Validation (PPMV) with MATLAB and Simulink Video Series
Part 1 Introduction
A three-step process for power plant model validation using MATLAB and Simulink
Part 2 Summary
Learn more on how to apply power plant model validation using online performance monitoring of grid events
Part 3 Manual Parameter Tuning
Gain deeper insight into response discrepancies through both VF replay and PQ replay Apply engineering judgement to adjust parameter settings
Part 4 Automated Parameter Sensitivity and Parameter Tuning
Compliment engineering judgement with automated parame-ter sensitivity to assess and rank the influence of system parameters on system response
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 2
Introduction
If you want to create a large-scale simulation of an electric grid that closely matches reality you gen-erally have to calibrate the individual components of that grid This process power plant model vali-dation (PPMV) can be performed with both offline step tests and online performance monitoring of grid events The main objective of PPMV is to
bull Find potential errors andor fixes in the model
bull Understand the sensitivity of parameters to potential model improvements
This task can be challenging particularly when required by technical regulations such as NERC Standard MOD-026
This paper covers a workflow for PPMV using MATLABreg and Simulinkreg with emphasis on online performance monitoring of grid events using phasor measurement unit (PMU) data It explores work-flows that include both manual adjustments and automated techniques A gas plant case study dem-onstrates how to
1 Replay measured data through your simulations
2 Gain insight into response discrepancies through both VF and PQ replay
3 Use engineering judgement and automated parameter sensitivity to assess and rank the influence of system parameters on system response
4 Fine-tune your system response using both manual adjustments and automated parameter estimation
The grid-connected gas plant model structure and grid event data were provided as part of the NASPINERC Technical Workshop on Model Verification Tools The model consists of the following components
bull Round rotor generator
bull General rotating exciter
bull GGOV1
bull PSS2A
bull Step-up transformer and series inductor
Replay Measured Data Through Your Simulations
The core attribute of PPMV is to replay measured data through a simulation model For offline step-tests data replay involves exciting the plant model through control signal measurements For online performance monitoring data replay involves exciting the plant model through physical measure-ments of voltage or current at the grid point-of-connection It is common to consider voltage (V) fre-quency (F) active-power (P) and reactive-power (Q) as the four measurements necessary to perform PPMV
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 3
Two data replay paradigms are available with these four measurements
1 PQ replay and compare simulated VF with measured VF
2 VF replay and compare simulated PQ with measured PQ
Figure 1 shows an example of the two data replay paradigms In this case the simulation model is cor-rupt and requires parameter modification A natural question that arises is which paradigm should be chosen to perform PPMV To obtain the most insight into system response and establish the most robust parameter estimation workflow the answer is both
Figure 1 VF replay with PQ comparison (top) and PQ replay with VF comparison (bottom)
Gain Insight into Response Discrepancies Through Both VF and PQ Replay
The first stage in PPMV is to gain insights into response discrepancies through manual adjustment of parameters You may have a list of ldquogo-tordquo parameters to adjust based on previous experience and can also gain additional insights through observing the attributes of a response discrepancy which can point to certain parameters requiring adjustment Figure 2 shows the results of PQ replay for the gas plant example As the simulated VF response does not match the measured VF response the model is corrupt and requires parameter adjustment The increased magnitude of frequency oscillation on the simulation indicates that generator inertia is too low Figure 3 shows the results of PQ replay with generator inertia increased The simulated oscillation now better fits the measured response
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 4
Figure 2 PQ replay for corrupt gas plant model
Figure 3 PQ replay with generator inertia increased
Figure 4 shows the results of VF replay for the gas plant example The reactive power response shows an underlying 180deg phase shift discrepancy in the simulated response This is indicative of the time-constants in two series washout filters in the power system stabilizer (PSS) requiring adjustment Figure 5 shows the results of VF replay with the PSS washout filter time-constants increased (Tw1 Tw2) The simulated reactive-power now better fits the measured response
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 5
Figure 4 VF replay for corrupt gas plant model
Figure 5 VF replay with PSS time-constants increased
Use Engineering Judgement and Automated Parameter Sensitivity to Assess and Rank the Influence of System Parameters on System Response
Further manual adjustments were made resulting in the response of Figure 6 for PQ replay and Figure 7 for VF replay Table 1 shows a comparison of parameter adjustments The results are certain-ly heading in the right direction and could be further improved manually but at this stage we will apply automated parameter sensitivity to assess and rank the influence of additional system parame-ters on system response
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 6
Figure 6 PQ replay for manually adjusted parameters
Figure 7 VF replay for manually adjusted parameters
Table 1 Manual parameter adjustments
Parameter Corrupt Value Manually Adjusted Value
pssTw1 10 60
pssTw2 10 60
pssKs1 150 250
govKturb 15 30
govKpgov 60 20
avrTf 50 10
genH 31 60
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 7
We use automated parameter sensitivity to assess parameters not evaluated during manual adjust-ments PQ replay and VF replay are used simultaneously for automated parameter sensitivity (PQVF replay) Figure 8 shows the model for PQ replay VF replay simply involves changing the data replay block indicated in red For PQVF replay the duplicate models share the same parameter values
Figure 8 Model for PQ replay (VF replay)
The objective function (f(x)) that is used to calculate sensitivity compares measured PQVF with sim-ulated PQVF The responses are normalized to mitigate scaling bias The objective function uses sum squared error (SSE)
Figure 9 shows a plot of the correlation of the plant parameters to the objective function While mag-nitude of correlation is certainly a factor for selecting a parameter sensitive parameters may be dis-counted based on engineering judgement For this case study we will select genTpd0 genTppd0 genXq and pssT4 for further consideration
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 8
Figure 9 Correlation of system parameter influence to objective function
Fine-Tune Your System Response Using Both Manual Adjustments and Automated Parameter Estimation
Following manual adjustment and automated parameter sensitivity you can apply automated param-eter estimation to fine-tune the response In this stage we have observed that solution convergence for PQVF replay is more robust and accurate than separate PQ replay or VF replay Parameter value ranges can be constrained in the automated parameter estimation task Both absolute and relative
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 9
constraints can be implemented Absolute constraints are typically adequate for control system parameters but relative constraints are required for generator parameters
Figure 10 shows a comparison of automated fine-tuning for both the manually selected parameters and the manually selected parameters in addition to the automatically selected parameters Bringing additional parameters to the parameter estimation task has improved the overall fit
Figure 10 Result of automated fine-tuning
It should be noted that the PPMV task should not end after automated fine-tuning You should assess the result and determine whether further manual adjustments can be made For example you can set parameters that do not change significantly back to their original values and compare responses If the result is comparable it may be more appropriate to stick with the minimum number of parameter changes Table 2 shows the final result of this case study where genTpd0 and genTppd0 did not change much during automated fine-tuning and were reset to their original values In this case the end result was slightly improved although this should not be regarded as a general outcome
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 1 0
Table 2 Case study results
Summary
In this paper we explored PPMV as applied to online performance monitoring of grid events using PMU data using a workflow that included both manual adjustments and automated techniques A gas plant case study demonstrated the following workflow steps
1 Replay measured data through your simulations
2 Gain insight into response discrepancies through both VF and PQ replay
3 Use engineering judgement and automated parameter sensitivity to assess and rank the influence of system parameters on system response
4 Fine-tune your system response using both manual adjustments and automated parameter estimation
With MATLAB and Simulink you can efficiently perform power plant model validation with auto-mated techniques The workflow provides insight and flexibility when addressing technical regula-tions such as NERC Standard MOD-26
Parameter Corrupt First-Pass Tuned
Second-Pass Tuned
Final
f(x) 9776 00872 00477 00472
pssKs1 150 2561 2564 2564
pssTw1 10 570 551 551
pssTw2 10 570 551 551
govKpgov 60 255 254 254
govKturb 15 30 30 30
avrTf 50 055 068 068
genH 31 60 60 60
genXq 13 13 177 177
genTppd0 025 025 023 025
genTpd0 50 50 505 50
pssT4 116 116 131 131
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 1 1
copy 2017 The MathWorks Inc MATLAB and Simulink are registered trademarks of The MathWorks Inc See mathworkscomtrademarks for a list of additional trademarks Other product or brand names may be trademarks or registered trademarks of their respective holders
93126v00 0517
Learn More About PPMV with MATLAB and Simulink
Power Plant Model Validation (PPMV) with MATLAB and Simulink Video Series
Part 1 Introduction
A three-step process for power plant model validation using MATLAB and Simulink
Part 2 Summary
Learn more on how to apply power plant model validation using online performance monitoring of grid events
Part 3 Manual Parameter Tuning
Gain deeper insight into response discrepancies through both VF replay and PQ replay Apply engineering judgement to adjust parameter settings
Part 4 Automated Parameter Sensitivity and Parameter Tuning
Compliment engineering judgement with automated parame-ter sensitivity to assess and rank the influence of system parameters on system response
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 3
Two data replay paradigms are available with these four measurements
1 PQ replay and compare simulated VF with measured VF
2 VF replay and compare simulated PQ with measured PQ
Figure 1 shows an example of the two data replay paradigms In this case the simulation model is cor-rupt and requires parameter modification A natural question that arises is which paradigm should be chosen to perform PPMV To obtain the most insight into system response and establish the most robust parameter estimation workflow the answer is both
Figure 1 VF replay with PQ comparison (top) and PQ replay with VF comparison (bottom)
Gain Insight into Response Discrepancies Through Both VF and PQ Replay
The first stage in PPMV is to gain insights into response discrepancies through manual adjustment of parameters You may have a list of ldquogo-tordquo parameters to adjust based on previous experience and can also gain additional insights through observing the attributes of a response discrepancy which can point to certain parameters requiring adjustment Figure 2 shows the results of PQ replay for the gas plant example As the simulated VF response does not match the measured VF response the model is corrupt and requires parameter adjustment The increased magnitude of frequency oscillation on the simulation indicates that generator inertia is too low Figure 3 shows the results of PQ replay with generator inertia increased The simulated oscillation now better fits the measured response
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 4
Figure 2 PQ replay for corrupt gas plant model
Figure 3 PQ replay with generator inertia increased
Figure 4 shows the results of VF replay for the gas plant example The reactive power response shows an underlying 180deg phase shift discrepancy in the simulated response This is indicative of the time-constants in two series washout filters in the power system stabilizer (PSS) requiring adjustment Figure 5 shows the results of VF replay with the PSS washout filter time-constants increased (Tw1 Tw2) The simulated reactive-power now better fits the measured response
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 5
Figure 4 VF replay for corrupt gas plant model
Figure 5 VF replay with PSS time-constants increased
Use Engineering Judgement and Automated Parameter Sensitivity to Assess and Rank the Influence of System Parameters on System Response
Further manual adjustments were made resulting in the response of Figure 6 for PQ replay and Figure 7 for VF replay Table 1 shows a comparison of parameter adjustments The results are certain-ly heading in the right direction and could be further improved manually but at this stage we will apply automated parameter sensitivity to assess and rank the influence of additional system parame-ters on system response
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 6
Figure 6 PQ replay for manually adjusted parameters
Figure 7 VF replay for manually adjusted parameters
Table 1 Manual parameter adjustments
Parameter Corrupt Value Manually Adjusted Value
pssTw1 10 60
pssTw2 10 60
pssKs1 150 250
govKturb 15 30
govKpgov 60 20
avrTf 50 10
genH 31 60
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 7
We use automated parameter sensitivity to assess parameters not evaluated during manual adjust-ments PQ replay and VF replay are used simultaneously for automated parameter sensitivity (PQVF replay) Figure 8 shows the model for PQ replay VF replay simply involves changing the data replay block indicated in red For PQVF replay the duplicate models share the same parameter values
Figure 8 Model for PQ replay (VF replay)
The objective function (f(x)) that is used to calculate sensitivity compares measured PQVF with sim-ulated PQVF The responses are normalized to mitigate scaling bias The objective function uses sum squared error (SSE)
Figure 9 shows a plot of the correlation of the plant parameters to the objective function While mag-nitude of correlation is certainly a factor for selecting a parameter sensitive parameters may be dis-counted based on engineering judgement For this case study we will select genTpd0 genTppd0 genXq and pssT4 for further consideration
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 8
Figure 9 Correlation of system parameter influence to objective function
Fine-Tune Your System Response Using Both Manual Adjustments and Automated Parameter Estimation
Following manual adjustment and automated parameter sensitivity you can apply automated param-eter estimation to fine-tune the response In this stage we have observed that solution convergence for PQVF replay is more robust and accurate than separate PQ replay or VF replay Parameter value ranges can be constrained in the automated parameter estimation task Both absolute and relative
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 9
constraints can be implemented Absolute constraints are typically adequate for control system parameters but relative constraints are required for generator parameters
Figure 10 shows a comparison of automated fine-tuning for both the manually selected parameters and the manually selected parameters in addition to the automatically selected parameters Bringing additional parameters to the parameter estimation task has improved the overall fit
Figure 10 Result of automated fine-tuning
It should be noted that the PPMV task should not end after automated fine-tuning You should assess the result and determine whether further manual adjustments can be made For example you can set parameters that do not change significantly back to their original values and compare responses If the result is comparable it may be more appropriate to stick with the minimum number of parameter changes Table 2 shows the final result of this case study where genTpd0 and genTppd0 did not change much during automated fine-tuning and were reset to their original values In this case the end result was slightly improved although this should not be regarded as a general outcome
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 1 0
Table 2 Case study results
Summary
In this paper we explored PPMV as applied to online performance monitoring of grid events using PMU data using a workflow that included both manual adjustments and automated techniques A gas plant case study demonstrated the following workflow steps
1 Replay measured data through your simulations
2 Gain insight into response discrepancies through both VF and PQ replay
3 Use engineering judgement and automated parameter sensitivity to assess and rank the influence of system parameters on system response
4 Fine-tune your system response using both manual adjustments and automated parameter estimation
With MATLAB and Simulink you can efficiently perform power plant model validation with auto-mated techniques The workflow provides insight and flexibility when addressing technical regula-tions such as NERC Standard MOD-26
Parameter Corrupt First-Pass Tuned
Second-Pass Tuned
Final
f(x) 9776 00872 00477 00472
pssKs1 150 2561 2564 2564
pssTw1 10 570 551 551
pssTw2 10 570 551 551
govKpgov 60 255 254 254
govKturb 15 30 30 30
avrTf 50 055 068 068
genH 31 60 60 60
genXq 13 13 177 177
genTppd0 025 025 023 025
genTpd0 50 50 505 50
pssT4 116 116 131 131
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 1 1
copy 2017 The MathWorks Inc MATLAB and Simulink are registered trademarks of The MathWorks Inc See mathworkscomtrademarks for a list of additional trademarks Other product or brand names may be trademarks or registered trademarks of their respective holders
93126v00 0517
Learn More About PPMV with MATLAB and Simulink
Power Plant Model Validation (PPMV) with MATLAB and Simulink Video Series
Part 1 Introduction
A three-step process for power plant model validation using MATLAB and Simulink
Part 2 Summary
Learn more on how to apply power plant model validation using online performance monitoring of grid events
Part 3 Manual Parameter Tuning
Gain deeper insight into response discrepancies through both VF replay and PQ replay Apply engineering judgement to adjust parameter settings
Part 4 Automated Parameter Sensitivity and Parameter Tuning
Compliment engineering judgement with automated parame-ter sensitivity to assess and rank the influence of system parameters on system response
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 4
Figure 2 PQ replay for corrupt gas plant model
Figure 3 PQ replay with generator inertia increased
Figure 4 shows the results of VF replay for the gas plant example The reactive power response shows an underlying 180deg phase shift discrepancy in the simulated response This is indicative of the time-constants in two series washout filters in the power system stabilizer (PSS) requiring adjustment Figure 5 shows the results of VF replay with the PSS washout filter time-constants increased (Tw1 Tw2) The simulated reactive-power now better fits the measured response
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 5
Figure 4 VF replay for corrupt gas plant model
Figure 5 VF replay with PSS time-constants increased
Use Engineering Judgement and Automated Parameter Sensitivity to Assess and Rank the Influence of System Parameters on System Response
Further manual adjustments were made resulting in the response of Figure 6 for PQ replay and Figure 7 for VF replay Table 1 shows a comparison of parameter adjustments The results are certain-ly heading in the right direction and could be further improved manually but at this stage we will apply automated parameter sensitivity to assess and rank the influence of additional system parame-ters on system response
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 6
Figure 6 PQ replay for manually adjusted parameters
Figure 7 VF replay for manually adjusted parameters
Table 1 Manual parameter adjustments
Parameter Corrupt Value Manually Adjusted Value
pssTw1 10 60
pssTw2 10 60
pssKs1 150 250
govKturb 15 30
govKpgov 60 20
avrTf 50 10
genH 31 60
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 7
We use automated parameter sensitivity to assess parameters not evaluated during manual adjust-ments PQ replay and VF replay are used simultaneously for automated parameter sensitivity (PQVF replay) Figure 8 shows the model for PQ replay VF replay simply involves changing the data replay block indicated in red For PQVF replay the duplicate models share the same parameter values
Figure 8 Model for PQ replay (VF replay)
The objective function (f(x)) that is used to calculate sensitivity compares measured PQVF with sim-ulated PQVF The responses are normalized to mitigate scaling bias The objective function uses sum squared error (SSE)
Figure 9 shows a plot of the correlation of the plant parameters to the objective function While mag-nitude of correlation is certainly a factor for selecting a parameter sensitive parameters may be dis-counted based on engineering judgement For this case study we will select genTpd0 genTppd0 genXq and pssT4 for further consideration
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 8
Figure 9 Correlation of system parameter influence to objective function
Fine-Tune Your System Response Using Both Manual Adjustments and Automated Parameter Estimation
Following manual adjustment and automated parameter sensitivity you can apply automated param-eter estimation to fine-tune the response In this stage we have observed that solution convergence for PQVF replay is more robust and accurate than separate PQ replay or VF replay Parameter value ranges can be constrained in the automated parameter estimation task Both absolute and relative
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 9
constraints can be implemented Absolute constraints are typically adequate for control system parameters but relative constraints are required for generator parameters
Figure 10 shows a comparison of automated fine-tuning for both the manually selected parameters and the manually selected parameters in addition to the automatically selected parameters Bringing additional parameters to the parameter estimation task has improved the overall fit
Figure 10 Result of automated fine-tuning
It should be noted that the PPMV task should not end after automated fine-tuning You should assess the result and determine whether further manual adjustments can be made For example you can set parameters that do not change significantly back to their original values and compare responses If the result is comparable it may be more appropriate to stick with the minimum number of parameter changes Table 2 shows the final result of this case study where genTpd0 and genTppd0 did not change much during automated fine-tuning and were reset to their original values In this case the end result was slightly improved although this should not be regarded as a general outcome
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 1 0
Table 2 Case study results
Summary
In this paper we explored PPMV as applied to online performance monitoring of grid events using PMU data using a workflow that included both manual adjustments and automated techniques A gas plant case study demonstrated the following workflow steps
1 Replay measured data through your simulations
2 Gain insight into response discrepancies through both VF and PQ replay
3 Use engineering judgement and automated parameter sensitivity to assess and rank the influence of system parameters on system response
4 Fine-tune your system response using both manual adjustments and automated parameter estimation
With MATLAB and Simulink you can efficiently perform power plant model validation with auto-mated techniques The workflow provides insight and flexibility when addressing technical regula-tions such as NERC Standard MOD-26
Parameter Corrupt First-Pass Tuned
Second-Pass Tuned
Final
f(x) 9776 00872 00477 00472
pssKs1 150 2561 2564 2564
pssTw1 10 570 551 551
pssTw2 10 570 551 551
govKpgov 60 255 254 254
govKturb 15 30 30 30
avrTf 50 055 068 068
genH 31 60 60 60
genXq 13 13 177 177
genTppd0 025 025 023 025
genTpd0 50 50 505 50
pssT4 116 116 131 131
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 1 1
copy 2017 The MathWorks Inc MATLAB and Simulink are registered trademarks of The MathWorks Inc See mathworkscomtrademarks for a list of additional trademarks Other product or brand names may be trademarks or registered trademarks of their respective holders
93126v00 0517
Learn More About PPMV with MATLAB and Simulink
Power Plant Model Validation (PPMV) with MATLAB and Simulink Video Series
Part 1 Introduction
A three-step process for power plant model validation using MATLAB and Simulink
Part 2 Summary
Learn more on how to apply power plant model validation using online performance monitoring of grid events
Part 3 Manual Parameter Tuning
Gain deeper insight into response discrepancies through both VF replay and PQ replay Apply engineering judgement to adjust parameter settings
Part 4 Automated Parameter Sensitivity and Parameter Tuning
Compliment engineering judgement with automated parame-ter sensitivity to assess and rank the influence of system parameters on system response
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 5
Figure 4 VF replay for corrupt gas plant model
Figure 5 VF replay with PSS time-constants increased
Use Engineering Judgement and Automated Parameter Sensitivity to Assess and Rank the Influence of System Parameters on System Response
Further manual adjustments were made resulting in the response of Figure 6 for PQ replay and Figure 7 for VF replay Table 1 shows a comparison of parameter adjustments The results are certain-ly heading in the right direction and could be further improved manually but at this stage we will apply automated parameter sensitivity to assess and rank the influence of additional system parame-ters on system response
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 6
Figure 6 PQ replay for manually adjusted parameters
Figure 7 VF replay for manually adjusted parameters
Table 1 Manual parameter adjustments
Parameter Corrupt Value Manually Adjusted Value
pssTw1 10 60
pssTw2 10 60
pssKs1 150 250
govKturb 15 30
govKpgov 60 20
avrTf 50 10
genH 31 60
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 7
We use automated parameter sensitivity to assess parameters not evaluated during manual adjust-ments PQ replay and VF replay are used simultaneously for automated parameter sensitivity (PQVF replay) Figure 8 shows the model for PQ replay VF replay simply involves changing the data replay block indicated in red For PQVF replay the duplicate models share the same parameter values
Figure 8 Model for PQ replay (VF replay)
The objective function (f(x)) that is used to calculate sensitivity compares measured PQVF with sim-ulated PQVF The responses are normalized to mitigate scaling bias The objective function uses sum squared error (SSE)
Figure 9 shows a plot of the correlation of the plant parameters to the objective function While mag-nitude of correlation is certainly a factor for selecting a parameter sensitive parameters may be dis-counted based on engineering judgement For this case study we will select genTpd0 genTppd0 genXq and pssT4 for further consideration
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 8
Figure 9 Correlation of system parameter influence to objective function
Fine-Tune Your System Response Using Both Manual Adjustments and Automated Parameter Estimation
Following manual adjustment and automated parameter sensitivity you can apply automated param-eter estimation to fine-tune the response In this stage we have observed that solution convergence for PQVF replay is more robust and accurate than separate PQ replay or VF replay Parameter value ranges can be constrained in the automated parameter estimation task Both absolute and relative
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 9
constraints can be implemented Absolute constraints are typically adequate for control system parameters but relative constraints are required for generator parameters
Figure 10 shows a comparison of automated fine-tuning for both the manually selected parameters and the manually selected parameters in addition to the automatically selected parameters Bringing additional parameters to the parameter estimation task has improved the overall fit
Figure 10 Result of automated fine-tuning
It should be noted that the PPMV task should not end after automated fine-tuning You should assess the result and determine whether further manual adjustments can be made For example you can set parameters that do not change significantly back to their original values and compare responses If the result is comparable it may be more appropriate to stick with the minimum number of parameter changes Table 2 shows the final result of this case study where genTpd0 and genTppd0 did not change much during automated fine-tuning and were reset to their original values In this case the end result was slightly improved although this should not be regarded as a general outcome
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 1 0
Table 2 Case study results
Summary
In this paper we explored PPMV as applied to online performance monitoring of grid events using PMU data using a workflow that included both manual adjustments and automated techniques A gas plant case study demonstrated the following workflow steps
1 Replay measured data through your simulations
2 Gain insight into response discrepancies through both VF and PQ replay
3 Use engineering judgement and automated parameter sensitivity to assess and rank the influence of system parameters on system response
4 Fine-tune your system response using both manual adjustments and automated parameter estimation
With MATLAB and Simulink you can efficiently perform power plant model validation with auto-mated techniques The workflow provides insight and flexibility when addressing technical regula-tions such as NERC Standard MOD-26
Parameter Corrupt First-Pass Tuned
Second-Pass Tuned
Final
f(x) 9776 00872 00477 00472
pssKs1 150 2561 2564 2564
pssTw1 10 570 551 551
pssTw2 10 570 551 551
govKpgov 60 255 254 254
govKturb 15 30 30 30
avrTf 50 055 068 068
genH 31 60 60 60
genXq 13 13 177 177
genTppd0 025 025 023 025
genTpd0 50 50 505 50
pssT4 116 116 131 131
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 1 1
copy 2017 The MathWorks Inc MATLAB and Simulink are registered trademarks of The MathWorks Inc See mathworkscomtrademarks for a list of additional trademarks Other product or brand names may be trademarks or registered trademarks of their respective holders
93126v00 0517
Learn More About PPMV with MATLAB and Simulink
Power Plant Model Validation (PPMV) with MATLAB and Simulink Video Series
Part 1 Introduction
A three-step process for power plant model validation using MATLAB and Simulink
Part 2 Summary
Learn more on how to apply power plant model validation using online performance monitoring of grid events
Part 3 Manual Parameter Tuning
Gain deeper insight into response discrepancies through both VF replay and PQ replay Apply engineering judgement to adjust parameter settings
Part 4 Automated Parameter Sensitivity and Parameter Tuning
Compliment engineering judgement with automated parame-ter sensitivity to assess and rank the influence of system parameters on system response
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 6
Figure 6 PQ replay for manually adjusted parameters
Figure 7 VF replay for manually adjusted parameters
Table 1 Manual parameter adjustments
Parameter Corrupt Value Manually Adjusted Value
pssTw1 10 60
pssTw2 10 60
pssKs1 150 250
govKturb 15 30
govKpgov 60 20
avrTf 50 10
genH 31 60
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 7
We use automated parameter sensitivity to assess parameters not evaluated during manual adjust-ments PQ replay and VF replay are used simultaneously for automated parameter sensitivity (PQVF replay) Figure 8 shows the model for PQ replay VF replay simply involves changing the data replay block indicated in red For PQVF replay the duplicate models share the same parameter values
Figure 8 Model for PQ replay (VF replay)
The objective function (f(x)) that is used to calculate sensitivity compares measured PQVF with sim-ulated PQVF The responses are normalized to mitigate scaling bias The objective function uses sum squared error (SSE)
Figure 9 shows a plot of the correlation of the plant parameters to the objective function While mag-nitude of correlation is certainly a factor for selecting a parameter sensitive parameters may be dis-counted based on engineering judgement For this case study we will select genTpd0 genTppd0 genXq and pssT4 for further consideration
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 8
Figure 9 Correlation of system parameter influence to objective function
Fine-Tune Your System Response Using Both Manual Adjustments and Automated Parameter Estimation
Following manual adjustment and automated parameter sensitivity you can apply automated param-eter estimation to fine-tune the response In this stage we have observed that solution convergence for PQVF replay is more robust and accurate than separate PQ replay or VF replay Parameter value ranges can be constrained in the automated parameter estimation task Both absolute and relative
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 9
constraints can be implemented Absolute constraints are typically adequate for control system parameters but relative constraints are required for generator parameters
Figure 10 shows a comparison of automated fine-tuning for both the manually selected parameters and the manually selected parameters in addition to the automatically selected parameters Bringing additional parameters to the parameter estimation task has improved the overall fit
Figure 10 Result of automated fine-tuning
It should be noted that the PPMV task should not end after automated fine-tuning You should assess the result and determine whether further manual adjustments can be made For example you can set parameters that do not change significantly back to their original values and compare responses If the result is comparable it may be more appropriate to stick with the minimum number of parameter changes Table 2 shows the final result of this case study where genTpd0 and genTppd0 did not change much during automated fine-tuning and were reset to their original values In this case the end result was slightly improved although this should not be regarded as a general outcome
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 1 0
Table 2 Case study results
Summary
In this paper we explored PPMV as applied to online performance monitoring of grid events using PMU data using a workflow that included both manual adjustments and automated techniques A gas plant case study demonstrated the following workflow steps
1 Replay measured data through your simulations
2 Gain insight into response discrepancies through both VF and PQ replay
3 Use engineering judgement and automated parameter sensitivity to assess and rank the influence of system parameters on system response
4 Fine-tune your system response using both manual adjustments and automated parameter estimation
With MATLAB and Simulink you can efficiently perform power plant model validation with auto-mated techniques The workflow provides insight and flexibility when addressing technical regula-tions such as NERC Standard MOD-26
Parameter Corrupt First-Pass Tuned
Second-Pass Tuned
Final
f(x) 9776 00872 00477 00472
pssKs1 150 2561 2564 2564
pssTw1 10 570 551 551
pssTw2 10 570 551 551
govKpgov 60 255 254 254
govKturb 15 30 30 30
avrTf 50 055 068 068
genH 31 60 60 60
genXq 13 13 177 177
genTppd0 025 025 023 025
genTpd0 50 50 505 50
pssT4 116 116 131 131
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 1 1
copy 2017 The MathWorks Inc MATLAB and Simulink are registered trademarks of The MathWorks Inc See mathworkscomtrademarks for a list of additional trademarks Other product or brand names may be trademarks or registered trademarks of their respective holders
93126v00 0517
Learn More About PPMV with MATLAB and Simulink
Power Plant Model Validation (PPMV) with MATLAB and Simulink Video Series
Part 1 Introduction
A three-step process for power plant model validation using MATLAB and Simulink
Part 2 Summary
Learn more on how to apply power plant model validation using online performance monitoring of grid events
Part 3 Manual Parameter Tuning
Gain deeper insight into response discrepancies through both VF replay and PQ replay Apply engineering judgement to adjust parameter settings
Part 4 Automated Parameter Sensitivity and Parameter Tuning
Compliment engineering judgement with automated parame-ter sensitivity to assess and rank the influence of system parameters on system response
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 7
We use automated parameter sensitivity to assess parameters not evaluated during manual adjust-ments PQ replay and VF replay are used simultaneously for automated parameter sensitivity (PQVF replay) Figure 8 shows the model for PQ replay VF replay simply involves changing the data replay block indicated in red For PQVF replay the duplicate models share the same parameter values
Figure 8 Model for PQ replay (VF replay)
The objective function (f(x)) that is used to calculate sensitivity compares measured PQVF with sim-ulated PQVF The responses are normalized to mitigate scaling bias The objective function uses sum squared error (SSE)
Figure 9 shows a plot of the correlation of the plant parameters to the objective function While mag-nitude of correlation is certainly a factor for selecting a parameter sensitive parameters may be dis-counted based on engineering judgement For this case study we will select genTpd0 genTppd0 genXq and pssT4 for further consideration
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 8
Figure 9 Correlation of system parameter influence to objective function
Fine-Tune Your System Response Using Both Manual Adjustments and Automated Parameter Estimation
Following manual adjustment and automated parameter sensitivity you can apply automated param-eter estimation to fine-tune the response In this stage we have observed that solution convergence for PQVF replay is more robust and accurate than separate PQ replay or VF replay Parameter value ranges can be constrained in the automated parameter estimation task Both absolute and relative
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 9
constraints can be implemented Absolute constraints are typically adequate for control system parameters but relative constraints are required for generator parameters
Figure 10 shows a comparison of automated fine-tuning for both the manually selected parameters and the manually selected parameters in addition to the automatically selected parameters Bringing additional parameters to the parameter estimation task has improved the overall fit
Figure 10 Result of automated fine-tuning
It should be noted that the PPMV task should not end after automated fine-tuning You should assess the result and determine whether further manual adjustments can be made For example you can set parameters that do not change significantly back to their original values and compare responses If the result is comparable it may be more appropriate to stick with the minimum number of parameter changes Table 2 shows the final result of this case study where genTpd0 and genTppd0 did not change much during automated fine-tuning and were reset to their original values In this case the end result was slightly improved although this should not be regarded as a general outcome
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 1 0
Table 2 Case study results
Summary
In this paper we explored PPMV as applied to online performance monitoring of grid events using PMU data using a workflow that included both manual adjustments and automated techniques A gas plant case study demonstrated the following workflow steps
1 Replay measured data through your simulations
2 Gain insight into response discrepancies through both VF and PQ replay
3 Use engineering judgement and automated parameter sensitivity to assess and rank the influence of system parameters on system response
4 Fine-tune your system response using both manual adjustments and automated parameter estimation
With MATLAB and Simulink you can efficiently perform power plant model validation with auto-mated techniques The workflow provides insight and flexibility when addressing technical regula-tions such as NERC Standard MOD-26
Parameter Corrupt First-Pass Tuned
Second-Pass Tuned
Final
f(x) 9776 00872 00477 00472
pssKs1 150 2561 2564 2564
pssTw1 10 570 551 551
pssTw2 10 570 551 551
govKpgov 60 255 254 254
govKturb 15 30 30 30
avrTf 50 055 068 068
genH 31 60 60 60
genXq 13 13 177 177
genTppd0 025 025 023 025
genTpd0 50 50 505 50
pssT4 116 116 131 131
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 1 1
copy 2017 The MathWorks Inc MATLAB and Simulink are registered trademarks of The MathWorks Inc See mathworkscomtrademarks for a list of additional trademarks Other product or brand names may be trademarks or registered trademarks of their respective holders
93126v00 0517
Learn More About PPMV with MATLAB and Simulink
Power Plant Model Validation (PPMV) with MATLAB and Simulink Video Series
Part 1 Introduction
A three-step process for power plant model validation using MATLAB and Simulink
Part 2 Summary
Learn more on how to apply power plant model validation using online performance monitoring of grid events
Part 3 Manual Parameter Tuning
Gain deeper insight into response discrepancies through both VF replay and PQ replay Apply engineering judgement to adjust parameter settings
Part 4 Automated Parameter Sensitivity and Parameter Tuning
Compliment engineering judgement with automated parame-ter sensitivity to assess and rank the influence of system parameters on system response
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 8
Figure 9 Correlation of system parameter influence to objective function
Fine-Tune Your System Response Using Both Manual Adjustments and Automated Parameter Estimation
Following manual adjustment and automated parameter sensitivity you can apply automated param-eter estimation to fine-tune the response In this stage we have observed that solution convergence for PQVF replay is more robust and accurate than separate PQ replay or VF replay Parameter value ranges can be constrained in the automated parameter estimation task Both absolute and relative
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 9
constraints can be implemented Absolute constraints are typically adequate for control system parameters but relative constraints are required for generator parameters
Figure 10 shows a comparison of automated fine-tuning for both the manually selected parameters and the manually selected parameters in addition to the automatically selected parameters Bringing additional parameters to the parameter estimation task has improved the overall fit
Figure 10 Result of automated fine-tuning
It should be noted that the PPMV task should not end after automated fine-tuning You should assess the result and determine whether further manual adjustments can be made For example you can set parameters that do not change significantly back to their original values and compare responses If the result is comparable it may be more appropriate to stick with the minimum number of parameter changes Table 2 shows the final result of this case study where genTpd0 and genTppd0 did not change much during automated fine-tuning and were reset to their original values In this case the end result was slightly improved although this should not be regarded as a general outcome
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 1 0
Table 2 Case study results
Summary
In this paper we explored PPMV as applied to online performance monitoring of grid events using PMU data using a workflow that included both manual adjustments and automated techniques A gas plant case study demonstrated the following workflow steps
1 Replay measured data through your simulations
2 Gain insight into response discrepancies through both VF and PQ replay
3 Use engineering judgement and automated parameter sensitivity to assess and rank the influence of system parameters on system response
4 Fine-tune your system response using both manual adjustments and automated parameter estimation
With MATLAB and Simulink you can efficiently perform power plant model validation with auto-mated techniques The workflow provides insight and flexibility when addressing technical regula-tions such as NERC Standard MOD-26
Parameter Corrupt First-Pass Tuned
Second-Pass Tuned
Final
f(x) 9776 00872 00477 00472
pssKs1 150 2561 2564 2564
pssTw1 10 570 551 551
pssTw2 10 570 551 551
govKpgov 60 255 254 254
govKturb 15 30 30 30
avrTf 50 055 068 068
genH 31 60 60 60
genXq 13 13 177 177
genTppd0 025 025 023 025
genTpd0 50 50 505 50
pssT4 116 116 131 131
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 1 1
copy 2017 The MathWorks Inc MATLAB and Simulink are registered trademarks of The MathWorks Inc See mathworkscomtrademarks for a list of additional trademarks Other product or brand names may be trademarks or registered trademarks of their respective holders
93126v00 0517
Learn More About PPMV with MATLAB and Simulink
Power Plant Model Validation (PPMV) with MATLAB and Simulink Video Series
Part 1 Introduction
A three-step process for power plant model validation using MATLAB and Simulink
Part 2 Summary
Learn more on how to apply power plant model validation using online performance monitoring of grid events
Part 3 Manual Parameter Tuning
Gain deeper insight into response discrepancies through both VF replay and PQ replay Apply engineering judgement to adjust parameter settings
Part 4 Automated Parameter Sensitivity and Parameter Tuning
Compliment engineering judgement with automated parame-ter sensitivity to assess and rank the influence of system parameters on system response
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 9
constraints can be implemented Absolute constraints are typically adequate for control system parameters but relative constraints are required for generator parameters
Figure 10 shows a comparison of automated fine-tuning for both the manually selected parameters and the manually selected parameters in addition to the automatically selected parameters Bringing additional parameters to the parameter estimation task has improved the overall fit
Figure 10 Result of automated fine-tuning
It should be noted that the PPMV task should not end after automated fine-tuning You should assess the result and determine whether further manual adjustments can be made For example you can set parameters that do not change significantly back to their original values and compare responses If the result is comparable it may be more appropriate to stick with the minimum number of parameter changes Table 2 shows the final result of this case study where genTpd0 and genTppd0 did not change much during automated fine-tuning and were reset to their original values In this case the end result was slightly improved although this should not be regarded as a general outcome
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 1 0
Table 2 Case study results
Summary
In this paper we explored PPMV as applied to online performance monitoring of grid events using PMU data using a workflow that included both manual adjustments and automated techniques A gas plant case study demonstrated the following workflow steps
1 Replay measured data through your simulations
2 Gain insight into response discrepancies through both VF and PQ replay
3 Use engineering judgement and automated parameter sensitivity to assess and rank the influence of system parameters on system response
4 Fine-tune your system response using both manual adjustments and automated parameter estimation
With MATLAB and Simulink you can efficiently perform power plant model validation with auto-mated techniques The workflow provides insight and flexibility when addressing technical regula-tions such as NERC Standard MOD-26
Parameter Corrupt First-Pass Tuned
Second-Pass Tuned
Final
f(x) 9776 00872 00477 00472
pssKs1 150 2561 2564 2564
pssTw1 10 570 551 551
pssTw2 10 570 551 551
govKpgov 60 255 254 254
govKturb 15 30 30 30
avrTf 50 055 068 068
genH 31 60 60 60
genXq 13 13 177 177
genTppd0 025 025 023 025
genTpd0 50 50 505 50
pssT4 116 116 131 131
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 1 1
copy 2017 The MathWorks Inc MATLAB and Simulink are registered trademarks of The MathWorks Inc See mathworkscomtrademarks for a list of additional trademarks Other product or brand names may be trademarks or registered trademarks of their respective holders
93126v00 0517
Learn More About PPMV with MATLAB and Simulink
Power Plant Model Validation (PPMV) with MATLAB and Simulink Video Series
Part 1 Introduction
A three-step process for power plant model validation using MATLAB and Simulink
Part 2 Summary
Learn more on how to apply power plant model validation using online performance monitoring of grid events
Part 3 Manual Parameter Tuning
Gain deeper insight into response discrepancies through both VF replay and PQ replay Apply engineering judgement to adjust parameter settings
Part 4 Automated Parameter Sensitivity and Parameter Tuning
Compliment engineering judgement with automated parame-ter sensitivity to assess and rank the influence of system parameters on system response
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 1 0
Table 2 Case study results
Summary
In this paper we explored PPMV as applied to online performance monitoring of grid events using PMU data using a workflow that included both manual adjustments and automated techniques A gas plant case study demonstrated the following workflow steps
1 Replay measured data through your simulations
2 Gain insight into response discrepancies through both VF and PQ replay
3 Use engineering judgement and automated parameter sensitivity to assess and rank the influence of system parameters on system response
4 Fine-tune your system response using both manual adjustments and automated parameter estimation
With MATLAB and Simulink you can efficiently perform power plant model validation with auto-mated techniques The workflow provides insight and flexibility when addressing technical regula-tions such as NERC Standard MOD-26
Parameter Corrupt First-Pass Tuned
Second-Pass Tuned
Final
f(x) 9776 00872 00477 00472
pssKs1 150 2561 2564 2564
pssTw1 10 570 551 551
pssTw2 10 570 551 551
govKpgov 60 255 254 254
govKturb 15 30 30 30
avrTf 50 055 068 068
genH 31 60 60 60
genXq 13 13 177 177
genTppd0 025 025 023 025
genTpd0 50 50 505 50
pssT4 116 116 131 131
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 1 1
copy 2017 The MathWorks Inc MATLAB and Simulink are registered trademarks of The MathWorks Inc See mathworkscomtrademarks for a list of additional trademarks Other product or brand names may be trademarks or registered trademarks of their respective holders
93126v00 0517
Learn More About PPMV with MATLAB and Simulink
Power Plant Model Validation (PPMV) with MATLAB and Simulink Video Series
Part 1 Introduction
A three-step process for power plant model validation using MATLAB and Simulink
Part 2 Summary
Learn more on how to apply power plant model validation using online performance monitoring of grid events
Part 3 Manual Parameter Tuning
Gain deeper insight into response discrepancies through both VF replay and PQ replay Apply engineering judgement to adjust parameter settings
Part 4 Automated Parameter Sensitivity and Parameter Tuning
Compliment engineering judgement with automated parame-ter sensitivity to assess and rank the influence of system parameters on system response
Power Plant Model Validation (PPMV) with MATLAB and Simulink
W H I T E PA P E R | 1 1
copy 2017 The MathWorks Inc MATLAB and Simulink are registered trademarks of The MathWorks Inc See mathworkscomtrademarks for a list of additional trademarks Other product or brand names may be trademarks or registered trademarks of their respective holders
93126v00 0517
Learn More About PPMV with MATLAB and Simulink
Power Plant Model Validation (PPMV) with MATLAB and Simulink Video Series
Part 1 Introduction
A three-step process for power plant model validation using MATLAB and Simulink
Part 2 Summary
Learn more on how to apply power plant model validation using online performance monitoring of grid events
Part 3 Manual Parameter Tuning
Gain deeper insight into response discrepancies through both VF replay and PQ replay Apply engineering judgement to adjust parameter settings
Part 4 Automated Parameter Sensitivity and Parameter Tuning
Compliment engineering judgement with automated parame-ter sensitivity to assess and rank the influence of system parameters on system response